Mark has been under a lot of scrutiny lately. All the privacy scandals has forced Facebook to rethink how they manage and use its members data. The results are turning into greater privacy control for end users, which is a really great thing. The internet needs more ways for users to control their own data, privately.

The backlash of these changes are having a negative impact for advertisers. Businesses rely on advanced targeting options within Facebook’s ad platform to get super specific on the users they want to show ads to. This has helped companies of all sizes be very efficient with their ad dollars. Now that Facebook is starting to put greater restrictions on their targeting options, advertisers are going to have to find a new way to get hyper specific with their targeting.

The good news? Conversion ads. Facebook has rolled out a newer ad objective called Conversion Ads that uses machine learning to help understand ideal target audiences and helping serve ads to the people that are most likely to convert. We’ve been playing with these ads for some time now, and have built a solid strategy around how to best utilize these. The results we have seen have been incredible.

In comparison to regular site traffic ads, we’ve seen conversion ads result in:

  • Higher time on-site
  • Lower bounce rates
  • Higher conversion rates
  • All while keeping the cost per click stable

Here’s how it works.

Website Conversions

First, you’ll want to establish a list of actions a user can take on your website that you would find to be valuable to your business. Common answers for most businesses could be: form submissions, phone number clicks, PDF downloads, address clicks to get directions, etc. Your list could look similar, but likely has some actions specific to your business goals. These actions are what signal that the quality of the traffic is high: these are the people you want on your site because they are the ones taking actions that will turn them into customers. They have “high intent”. We call a website session that takes one of these desirable actions a “conversion”.


Now that you have your list of conversion actions, you’ll want to start tracking them. The first step is to get them into your Google Analytics tracking system. Google Tag Manager is a powerful tool to help make this happen. Setup the properly triggers and tags to track all the conversion events that make sense for your business.

Once you have these events firing into Google Analytics, you’ll want to do the same into Facebook. With these conversion ads, Facebook now lets you tell it when a “conversion” takes place on your site. That way, Facebook can see what users are coming over to your site and taking desirable actions. The Facebook platform has some pre-built conversion actions you can use, but you can also create your own. With tracking in place, you’ll be able to see your conversions right inside Google Analytics but also inside Facebook’s Analytics.

Machine Learning Targeting

As conversions take place, Facebook is going track each user that converts and it’s going to look at all the data it has stored on that individual. The more conversions that it tracks, the better, because it will start to see patterns of similar attributes between all the users that convert. The machine learning algorithms at Facebook put these patterns together to start building out an ideal target audience. Now that it can understand what types of attributes users have that usually convert, it can go out and target additional users that have those similar attributes.

Historically, we’ve picked out the hyper targeting options using all the available options in the advertising dashboard to try and find people that will convert. Since a number of these options are change or going away all together, we can instead let Facebook establish our hyper-targeting by using machine learning to track conversions, learn about the users that convert, and let it automatically go find more people similar to that those “converters”.

Here’s an example of some data we’ve seen from a recent testing campaign that ran traditional Facebook ads side-by-side with Facebook Conversion Ads.

Ad Type Bounce Rate Pages per Session Avg Session Duration Goal Conversion Rate
Traditional Ad Targeting 68% 1.85 0:46 0.93%
Conversion Ad Targeting 53% 2.46 1:13 2.50%

Not only are the on-site behaviors showing significant improvement indicating higher quality traffic, but the conversion rates are over double. This is taking Facebook advertising to the next level. There is some technical setup on the backend to make this all happen, but the results are well worth the time investment to get it rolling.

If you’re looking for help on taking advantage of Facebook’s Conversion Ads, or just want to see better performance from your online marketing, we’re here for you.

We bring brands to life online to maximize their digital footprint. Contact us today to learn more about working with us.